| { |
| "best_metric": 0.8776, |
| "best_model_checkpoint": "./resnet_finetuned_models_dataset/CIFAR100/50_from_100/microsoft_resnet-101/model_idx_0032/checkpoints/checkpoint-2997", |
| "epoch": 9.0, |
| "eval_steps": 500, |
| "global_step": 2997, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 1.0, |
| "grad_norm": 74.93437957763672, |
| "learning_rate": 6.788924172750679e-05, |
| "loss": 3.3058, |
| "step": 333 |
| }, |
| { |
| "epoch": 1.0, |
| "eval_accuracy": 0.6256, |
| "eval_loss": 1.7701362371444702, |
| "eval_runtime": 13.4341, |
| "eval_samples_per_second": 279.14, |
| "eval_steps_per_second": 4.392, |
| "step": 333 |
| }, |
| { |
| "epoch": 2.0, |
| "grad_norm": 65.4781723022461, |
| "learning_rate": 6.181155550916422e-05, |
| "loss": 1.0064, |
| "step": 666 |
| }, |
| { |
| "epoch": 2.0, |
| "eval_accuracy": 0.8114666666666667, |
| "eval_loss": 0.6770989298820496, |
| "eval_runtime": 13.3209, |
| "eval_samples_per_second": 281.512, |
| "eval_steps_per_second": 4.429, |
| "step": 666 |
| }, |
| { |
| "epoch": 3.0, |
| "grad_norm": 122.15306091308594, |
| "learning_rate": 5.2499999999999995e-05, |
| "loss": 0.5517, |
| "step": 999 |
| }, |
| { |
| "epoch": 3.0, |
| "eval_accuracy": 0.8330666666666666, |
| "eval_loss": 0.5452556610107422, |
| "eval_runtime": 13.3285, |
| "eval_samples_per_second": 281.353, |
| "eval_steps_per_second": 4.427, |
| "step": 999 |
| }, |
| { |
| "epoch": 4.0, |
| "grad_norm": 88.65290069580078, |
| "learning_rate": 4.1077686218342566e-05, |
| "loss": 0.383, |
| "step": 1332 |
| }, |
| { |
| "epoch": 4.0, |
| "eval_accuracy": 0.8626666666666667, |
| "eval_loss": 0.46090593934059143, |
| "eval_runtime": 13.2665, |
| "eval_samples_per_second": 282.667, |
| "eval_steps_per_second": 4.447, |
| "step": 1332 |
| }, |
| { |
| "epoch": 5.0, |
| "grad_norm": 139.25636291503906, |
| "learning_rate": 2.8922313781657438e-05, |
| "loss": 0.2877, |
| "step": 1665 |
| }, |
| { |
| "epoch": 5.0, |
| "eval_accuracy": 0.8669333333333333, |
| "eval_loss": 0.44193699955940247, |
| "eval_runtime": 13.4104, |
| "eval_samples_per_second": 279.633, |
| "eval_steps_per_second": 4.4, |
| "step": 1665 |
| }, |
| { |
| "epoch": 6.0, |
| "grad_norm": 66.44093322753906, |
| "learning_rate": 1.7500000000000005e-05, |
| "loss": 0.2326, |
| "step": 1998 |
| }, |
| { |
| "epoch": 6.0, |
| "eval_accuracy": 0.872, |
| "eval_loss": 0.42644748091697693, |
| "eval_runtime": 13.2718, |
| "eval_samples_per_second": 282.554, |
| "eval_steps_per_second": 4.446, |
| "step": 1998 |
| }, |
| { |
| "epoch": 7.0, |
| "grad_norm": 242.61793518066406, |
| "learning_rate": 8.188444490835772e-06, |
| "loss": 0.2009, |
| "step": 2331 |
| }, |
| { |
| "epoch": 7.0, |
| "eval_accuracy": 0.8744, |
| "eval_loss": 0.4255537986755371, |
| "eval_runtime": 13.3942, |
| "eval_samples_per_second": 279.972, |
| "eval_steps_per_second": 4.405, |
| "step": 2331 |
| }, |
| { |
| "epoch": 8.0, |
| "grad_norm": 150.09632873535156, |
| "learning_rate": 2.1107582724932087e-06, |
| "loss": 0.1782, |
| "step": 2664 |
| }, |
| { |
| "epoch": 8.0, |
| "eval_accuracy": 0.876, |
| "eval_loss": 0.4164135456085205, |
| "eval_runtime": 13.271, |
| "eval_samples_per_second": 282.571, |
| "eval_steps_per_second": 4.446, |
| "step": 2664 |
| }, |
| { |
| "epoch": 9.0, |
| "grad_norm": 185.41868591308594, |
| "learning_rate": 0.0, |
| "loss": 0.1712, |
| "step": 2997 |
| }, |
| { |
| "epoch": 9.0, |
| "eval_accuracy": 0.8776, |
| "eval_loss": 0.4128022789955139, |
| "eval_runtime": 13.2439, |
| "eval_samples_per_second": 283.149, |
| "eval_steps_per_second": 4.455, |
| "step": 2997 |
| } |
| ], |
| "logging_steps": 500, |
| "max_steps": 2997, |
| "num_input_tokens_seen": 0, |
| "num_train_epochs": 9, |
| "save_steps": 500, |
| "stateful_callbacks": { |
| "TrainerControl": { |
| "args": { |
| "should_epoch_stop": false, |
| "should_evaluate": false, |
| "should_log": false, |
| "should_save": true, |
| "should_training_stop": true |
| }, |
| "attributes": {} |
| } |
| }, |
| "total_flos": 7.3587863155968e+18, |
| "train_batch_size": 64, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
